Can't connect to HA anymore | Home Assistant | Page 1

echo escarp Aug 20, 2025, 4:20 AM

#

Hi. I'm in a bit of a picke, my HA instance is supposed to be running, The observer page is running and shows everything in green. Yet, I can't connect to port 8123. It happened all of a sudden. This system has been running for years.

Previously I had been messing with the ZigBee2MQTT addons trying to change coordinator hardware. Changes I've now reverted.

I can ssh to the machine both on port 22222 and plain port 22 ; I see everything, I see updated logs etc. but no connection 🙁

Any suggestions on where to go next ?

sudden imp Aug 20, 2025, 4:32 AM

#

have you done a full reboot? Also check DNS/DHCP/NTP on network and try to login to HA with ip address instead of hostname

echo escarp Aug 20, 2025, 4:33 AM

#

I have. I am using an IP address.

sudden imp Aug 20, 2025, 4:33 AM

#

are you running reverse proxy or other dns stuff?

echo escarp Aug 20, 2025, 4:33 AM

#

when I ssh into the host, it tells me that it's running on port 8123, but nothing is listening there

#

right now I'm using directly the local IP address

#

I restored a backup that I thought was of last week, things have restarted, except I made a mistake I missed that it was a 2024 one 🙂

sudden imp Aug 20, 2025, 4:34 AM

#

oops

#

but at least its running 😉

echo escarp Aug 20, 2025, 4:35 AM

#

well, it's responding to port 8123

sudden imp Aug 20, 2025, 4:35 AM

#

clear cache browser, try again

echo escarp Aug 20, 2025, 4:35 AM

#

i don't know how long it will take for this restore to complete ; I started it from the CLI, it's been a while now

sudden imp Aug 20, 2025, 4:36 AM

#

depents on how big the backup is. would give it an hour to be sure.

echo escarp Aug 20, 2025, 4:37 AM

#

how are HA backups sorted in the CLI ? seems all over the place. The last one that showed was ``` - hass
name: "2024-08-06 03:00:00"
protected: false
size: 1172.72
size_bytes: 1229690880
slug: "43776708"
type: full

#

but the one above ``` - hass
name: "2025-05-10 03:00:00"
protected: false
size: 645.57
size_bytes: 676925440
slug: df957474
type: full

sudden imp Aug 20, 2025, 4:37 AM

#

my backup is 60MB, so be patient

#

how it is listed, i dont know, restore newest one?

echo escarp Aug 20, 2025, 4:39 AM

#

done now. I'm trying to determine which one is the latest, I have thousands of lines of backup and they aren't under any obvious ordered

#

I have one from this morning at 3AM ; it seems that I have over 2 years of daily full backup, not in any particular order !

echo escarp Aug 20, 2025, 4:58 AM

#

hmmmm, backup restore has completed but that shows after "Error: No Home Assistant Core response, assuming a fatal startup error"

#

system was definitely up and running this moring

sudden imp Aug 20, 2025, 5:05 AM

#

can you boot in to safe mode? if you dont care about historical data, you could delete the database. HA will create a fresh one.

echo escarp Aug 20, 2025, 5:05 AM

#

that's after rebooting the machine

Screenshot_2025-08-20_at_3.05.22_pm.heic

#

and with a backup from last night same as now

#

yet 2024 backup did restore

#

how do you boot into safe mode ?

#

losing historical data would be annoying, particularly the energy tracking

sudden imp Aug 20, 2025, 5:07 AM

#

CLI> type login and hit enter, then type ha core restart --safe-mode

#

if running HAOS

echo escarp Aug 20, 2025, 5:08 AM

#

it's a VM running HAOS

#

"Error: Another job is running for job group home_assistant_core"

sudden imp Aug 20, 2025, 5:08 AM

#

for core docker hass --safe-mode

echo escarp Aug 20, 2025, 5:08 AM

#

Processing... Done.

Error: Another job is running for job group home_assistant_core

sudden imp Aug 20, 2025, 5:09 AM

#

have you rebooted HOST too? Proxmox/Virtualbox etc.

echo escarp Aug 20, 2025, 5:09 AM

#

i've rebooted the machine yes

#

can do that again

sudden imp Aug 20, 2025, 5:10 AM

#

i would give it some time, if not responsive in 1 hour, then reboot complete host and re-run above commands

#

the error indicates that ha is doing "something"

#

you can validate configuration.yaml with command ha core check

echo escarp Aug 20, 2025, 5:13 AM

#

ok. ill wait a bit. weird that a 2024 backup would restore and boot, but a 2025 from yesterdya wouldn't

#

i saw some errors in the logs about my ip_bans having invalid entry. I manually entered some IP as a subnet it doesn't appear to like that

#

but I don't think that would have prevented things to boot

#

ha core check gives me "done" ; restart tells me that there's another job, but I don't think there's one

echo escarp Aug 20, 2025, 5:53 AM

#

I can see lines being added to home-assistant.log ; yet nothing listening to port 8123.

#

I could run the command ```# ha core restart --safe-mode
Processing... Done.

Error: No Home Assistant Core response, assuming a fatal startup error```

#

it took a while for "processing.... done" to appear, like 2+ minutes

#

if I delete home-assistant_v2.db will I need to redo absolutely everything or will just the history disappear ?

sudden imp Aug 20, 2025, 6:00 AM

#

just history.

echo escarp Aug 20, 2025, 6:01 AM

#

backup from last july 1st boots. So far backup up to August 13th do not

#

very puzzling

sudden imp Aug 20, 2025, 6:04 AM

#

would do july then and update..

echo escarp Aug 20, 2025, 6:05 AM

#

I got a battery installed on July 20th ; I've done a faire amount of work to create a new cards and the like to control it all 🙁

sudden imp Aug 20, 2025, 6:08 AM

#

have you checked your yaml?

#

you can validate configuration.yaml with command ha core check

echo escarp Aug 20, 2025, 6:13 AM

#

yes, this is all good.

#

August 12th back has restore fine.

sudden imp Aug 20, 2025, 6:14 AM

#

Superb

echo escarp Aug 20, 2025, 6:15 AM

#

i guess I will have to settle on that.

sudden imp Aug 20, 2025, 6:16 AM

#

Better than july 😉

echo escarp Aug 20, 2025, 6:16 AM

#

I have zero recollection where all those backups are stored 🙂

#

they aren't small either

sudden imp Aug 20, 2025, 6:17 AM

#

Maybe running out of space?

echo escarp Aug 20, 2025, 6:17 AM

#

about 1gig each, and I have over 2 years worth of them

#

IIRC, I used a smb share to my nas

sudden imp Aug 20, 2025, 6:17 AM

#

🤣

echo escarp Aug 20, 2025, 6:18 AM

#

I have 12x4TB disk on that machine, it's a TrueNAS where the HAOS VM is running

sudden imp Aug 20, 2025, 6:19 AM

#

Yeah, but its about the gigs you have assigned to vm though

echo escarp Aug 20, 2025, 6:19 AM

#

not if the backup are configured to use the SMB share

#

or maybe NFS

#

yep, 747GB of backup on that share. jeez

#

888 full backups

sudden imp Aug 20, 2025, 6:23 AM

#

wow lol

echo escarp Aug 20, 2025, 6:24 AM

#

That HA install controls everything in my house, not having it on would be a major bugger. wasted a day still 🙁

sudden imp Aug 20, 2025, 6:24 AM

#

using nabu casa, no worries about backups.. 😉

#

would not like to do my install again, i suggest you do a full backup next, when its uptodate and download backup to a pc.

#

in emergency, create new vm and install backup

echo escarp Aug 20, 2025, 6:26 AM

#

I do daily full backup, at 3AM

#

but for some reasons, recent ones don't boot

#

the only thing I've changed in the last 4 days is install 2025.8.2 and update Kiosk Mode

sudden imp Aug 20, 2025, 6:28 AM

#

me too, but did not break things

#

maybe do a harddisk check?

echo escarp Aug 20, 2025, 6:31 AM

#

it's a ZFS2 (RAID60) very unlikely to have things go wrong

#

maybe it was an addon ; how do I restore just the HA database ?

#

want to see if HA from earlier with latest database will work

#Can't connect to HA anymore